• No results found

The influence of perceived workload and job-insecurity on the amount of perceived job-stress among employees in the Dutch banking sector

N/A
N/A
Protected

Academic year: 2021

Share "The influence of perceived workload and job-insecurity on the amount of perceived job-stress among employees in the Dutch banking sector"

Copied!
56
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The influence of perceived workload and job-insecurity on the

amount of perceived job-stress among employees in the Dutch

banking sector

Name: Laura Maria Martje Hogenhout Student number: 10632298

Thesis supervisor: Nesrien Abu Ghazaleh BSc- programme: Economics & Business Specialization: Business Administration Date: 29-06-2016

(2)

Statement of Originality

This document is written by Laura Maria Martje Hogenhout who declares to take full

responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Abstract

Job-stress is widely recognized as a major problem for employees and their organizations. Solutions to this problem would be desired. Previous studies have shown that high levels of perceived workload and job-insecurity both lead to high levels of perceived job-stress among employees within several sectors. One of the sectors which is characterized by high levels of job-insecurity and workload, is the banking sector. However, the factors influencing job-stress among Dutch banking employees have only been investigated to a limited extent, and often before the financial crisis of 2008 took place. Furthermore, a potential moderating effect of job-insecurity on the relationship between workload and job-stress has, to our knowledge, not been investigated yet. In this study, it was expected that job-insecurity has a positive moderating effect. People who are highly insecure about retaining their current job, could be highly afraid of the consequences of not being able to cope with a very high workload, since this could indicate a low performance and can have a lay-off as a result. This study attempted to study the influence of the job-stressors workload and job-insecurity on the subsequent amount of job-stress among banking employees and investigated their combined effect as well. The goal was to in the end create more insight into the factors which lead to job-stress, to generate useful recommendations for managers and employees regarding ways to cope with, and reduce the levels of job-stressors and job-stress within this sector. Data for this study was gathered among 92 employees within the Dutch banking sector. First of all, it was found that an increase in workload leads to an increase in job-stress (H1). It is thus of significant importance that managers try to restrict the consequences of high workloads among their employees. In contrast, no significant main effect of insecurity on stress was found (H2). Furthermore, no moderating effect of job-insecurity on the relationship between workload and job-stress was found (H3).

(3)

Table of content

1. Introduction………4

2. Theoretical framework………...8

2.1. Job-stress……….8

2.2. The consequences of job-stress………...9

2.3. Reducing and preventing job-stress……….……….……….10

2.4. The variety of factors influencing job-stress……….……….11

2.5. Workload increases job-stress……….………...11

2.6. Reorganizations and workload……….……….13

2.7. Job-insecurity increases job-stress……….………...14

2.8. Reorganizations and job-insecurity...16

2.9 Moderating effect of job-insecurity on relationship between workload and job-stress.17 3. Method……….………...………..19 3.1. Research setting………...………….……….19 3.2. Data collection……….……….………...………..19 3.3. Research sample………19 3.4. Variables………...20 4. Results……….…...24

4.1. Correlations and reliabilities……….………...24

4.2. Regression analyses ……...………....…………...26

4.3. Regression results regarding hypotheses 1 & 2……….…………..…..28

4.4. Regression results regarding hypothesis 3……….…………30

5. Discussion……….…………....32

5.1. Findings………...….…...32

5.2. Theoretical implications and contributions to current literature………...………38

5.3. Practical implications………39

5.4. Limitations………....…….40

5.5. Directions for future research………....…………41

6. Conclusion………....………43

References………....…………44

(4)

4 1. Introduction

In today’s working environment, job-stress is a serious problem for both employees and

organizations (Boyd, Lewin & Sager, 2009). According to Dhankar (2015, p.134), work-related stress is ‘’a leading feature of modern life’’. It is evident that job-stress brings high costs. Work-related stress threatens the health of employees and leads to high costs for organizations (Mucci et al., 2015). Several studies have shown that employees who perceive high levels of job-stress incur high health costs (Ganster & Rosen, 2013). Next to direct health costs associated with stress, there are also costs concerned with lost working days as a result of job-stress. Chapman (2005) estimated that job-related strains cost organizations 300 billion dollars in the US only. These costs are unnecessary high. Therefore, it is of significant importance to find out which factors lead to perceived job-stress among employees and how these factors can be attacked by employers.

One of the sectors in which dramatic problems of stress are widely recognized is the banking sector. It was found in previous studies that employees in the banking sector perceive high levels of job-stress compared to employees in other sectors (van Ruysseveldt, Manshoven, de Witte & Bundervoet, 2003) and burnouts and stress were found to be the most important factors in explaining absenteeism in the Dutch banking sector (Bondgenoten, 2012). Employees with burnouts and stress cost organizations on average 36000 euros per year (Bondgenoten, 2012). Another research, conducted by the Gelfond Group, studied employees of the European Central Bank and a remarkable finding was that 37 percent of the participants experienced a decrease in their health due to stress at work (‘’Medewerkers ECB kampen met hoge werkdruk’’, 2015). This asks for further research in the direction of finding ways for managers to attack this problem of job-stress in their organization. This study seeks to contribute to theory by studying job-stress and related factors in a setting which is not often investigated until today: the Dutch banking sector. Job-stress is for example often studied in the health sector (Goldenberg and Waddell, 1990), however this phenomenon is less often investigated in the banking sector. Another theoretical contribution of this research, is that factors underlying job-stress will be combined in new ways. This means that factors which are often studied as independent factors, will be tested on the presence of a potential moderation effect.

(5)

5

Job-stress is an often studied phenomenon and different definitions can be found. A general definition which will be applied in this study is the definition from the National Institute of Safety and Health (1999), who defines job-stress as: ‘’ the harmful physical and emotional responses that occur when the requirements of the job do not match the capabilities, resources, or needs of the worker’’. Many factors in the working-environment have been investigated as job-stressors. Two of them are often cited as having a large influence on the amount of perceived job-stress while they also can be expected to be related to each other. These two factors are the amount of perceived job-insecurity and the perceived workload. These two factors will be further explained in detail.

In 2011, the American Psychological Association conducted a study on stress in the workplace and the high number of individuals suffering from job-stress was worrying. One of the main findings was the fact that a heavy workload was a significant source of the high experienced level of job-stress (Harris, 2011). Workload can be seen as the amount of work which has to be accomplished by an individual (Spector & Jex, 1998). Spector (1987) also found significant positive correlations between workload and anxiety, frustration, health symptoms, and job dissatisfaction.

Another factor often associated with job-stress is job-insecurity, which was also found in a study conducted by Harris in 2011. According to de Witte (1999), job-insecurity means that employees are afraid to lose their current job. Job-insecurity is often considered as a stress-related variable (Barling & Kelloway, 1996; Sonnentag & Frese, 2003). Those employees who report a high degree of insecurity in their work also report more strain, which is related to stress (Francis & Barling, 2005). Also in a study conducted by TSI Success Insights (2015), job-insecurity was mentioned as one of several main factors leading to stress at work. Job-job-insecurity often results from reorganizations in which jobs are being removed and employees are fired (Tijdens & van Klaveren, 2002). Job-insecurity belongs to the most important work stressors, however it is only to a limited extent clear to what extent job-insecurity predicts job-stress compared to other factors (Klandermans, van Vuuren & Jacobson, 1991; de Witte, 1999).

The banking sector has been the main victim of the financial crisis. Their balances were disturbed, their reputations were damaged and customers’ trust in the banking sector has declined as a result of the problems during the recession (‘’Bankencrisis’’, 2016). It is important for banks to recover from this situation, which could induce high pressure on their employees. For

(6)

6

example, van Ruysseveldt et al. (2003) made a comparison between workers in the banking sector and workers in other sectors and showed that workers in the banking sector experienced more stress than employees in other sectors. Mainly due to the financial crisis, worldwide it was estimated that around 600.000 employees within this sector have lost their job since 2008. Banks try to improve their financial position by reducing their costs. This is often done by

implementing reorganizations (‘’Banken schrappen ruim half miljoen banen sinds crisis, 2016). Van Ruysseveldt et al. (2003) found that employees in the banking sector perceived higher levels of job-insecurity than employees in other sectors. According to the NOS (2014), one out of three jobs at the four main players in the Netherlands disappeared as a result of downsizing efforts and reorganizations. The logical result of this downsizing is that the same amount of work has to be done with less employees within the same time and the job-security of these employees is expected to have decreased.

Next to a generally high job-insecurity within the banking sector, van Ruysseveldt et al. (2003) found that employees in the banking sector also experience a higher workload than workers in other sectors. They stated that this higher workload often explains the higher level of perceived job-stress. Van Ruysseveldt et al. (2003) emphasized the relation between a high perceived workload and job-stress. This makes it very interesting to look at the effects of a decreased job-security and an increase in workload on the subsequent perceived job-stress of the employees in this Dutch banking sector.

These factors, which seem to be highly present in the banking sector have only been investigated to a limited extent, most of the times in countries different from the Netherlands, as for example India and Belgium (Azad, 2014; van Ruysseveldt et al., 2003; Niharika and Kiran, 2014). Another fact which makes it especially interesting to look at these relationships, as already mentioned, is the recent financial crisis which could have higher job-insecurity and workloads as a logical result. Furthermore, it could be the case that the relationships between the factors job-insecurity and workload and the subsequent job-stress have changed during these years of financial crisis. This study will investigate these facets in the Dutch banking sector to get an updated, detailed insight into these relationships.

Besides these main effects of job-insecurity and workload, it could potentially be the case that job-insecurity plays a moderating role in the relationship between perceived workload and job-stress. Van Ruysseveldt, Cambré, Depickere and Adiele (2004) stated that employees in the

(7)

7

banking sector are forced to work very hard for a relative insecure job. Job-insecurity and workload are several times investigated in general and in the banking sector. However, a potential interaction effect of these two job-stressors has, for so far as we know, not been investigated yet. There is reason to think about a potential moderating effect. Ruysseveldt et al. (2004) stated that there is an interaction effect of workload and perceived control of employees together on their job-stress, which is in accordance with the job demand-control model of Karasek (Ruysseveldt et al., 2004). Job-insecurity is not included as potential moderator in the relationship between workload and job-stress in current studies. This is remarkable, since in fact, job-insecurity can be expected to have an interaction effect with workload on job-stress as well. Job-insecurity can be seen as another form of control or autonomy over retaining one’s job. It might be the case that when employees perceive their job as insecure, the effect of workload on job-stress becomes stronger. However, this potential relationship is a new idea which was generated during studying the current literature and previous studies.

This study will investigate these potential relations into more detail within the Dutch banking sector to eventually answer the overall research question: ‘’ How do job-insecurity and workload influence the amount of perceived job-stress in the Dutch banking sector?’’. Below, one can find the conceptual model which will be investigated within this study.

(8)

8

This paper will start with a comprehensive literature review, in which job-stress will be

explained into more detail. A general overview will be given of the different categories of factors which can lead to job-stress. Of all these variables, the variables included in this study and the expected relationships between these variables will be the main focus point and these will be discussed on the basis of current literature and previous studies. Thereafter, in the method section, it will be elaborated how the overall research question and the sub-questions will be answered. All details regarding data collection can be found in the method section. The method section will be followed by the results section in which the data will be analysed and the results will be disclosed. In the subsequent discussion section, these results will be critically

overthought and interpreted from various perspectives. Furthermore, the discussion will include implications of this study, limitations and directions for future research. To conclude, this paper will end with a conclusion in which the overall findings will be summarized.

2. Theoretical framework

2.1 Job-stress

Job-stress is an often discussed phenomenon and people are increasingly dealing with it in their work-life. According to the National Institute of Safety and Health (1999), job-stress can be described as ‘’ the harmful physical and emotional responses that occur when the requirements of the job do not match the capabilities, resources, or needs of the worker’’. Jamal (2005, p.225) defines job-stress as ‘’an individual’s reactions to the characteristics of the work environment that seem emotionally and physically threatening’’. The definition emphasizes a poor fit between the capabilities of an employee and the work environment (Jamal, 2005). In this work

environment, it could be the case that very high demands are placed under the responsibility of employees or that employees are not fully capable to handle a particular task on the job (Jamal, 1985). Overall, the greater the misfit between the demands placed upon the employee and his or her abilities, the higher stress-levels they will experience (Jamal, 2005). Stress is also often described as the internal response of an individual to a stressor. These stressors can be events or characteristics in the environment which impact employees (Jamal, 1985). The terms stress and strain are often used interchangeable. However, there is a clear distinction. Strain is related to stress and is the result of experiencing stress on a long term. Strain can be recognized as anxiety

(9)

9

and depression (Kayastha, Adhikary & Krishnamurthy, 2012). Often signs of strain are seen as indicators of being exposed to stress.

2.2 The consequences of job-stress

Job-stress has an impact on employees and their organizations. According to Ganster and Schaubroeck (1991) it is widely acknowledged that there is a causal relationship between work-stress and mental and physical disorders, and between work-work-stress and organizational outcomes such as absenteeism and productivity. These impacts on employees and their organizations are interrelated as will be explained below.

For individual employees, it is generally accepted that a long-term exposure to stressful job-demands can lead to several outcomes in the form of diseases (Ganster & Schaubroeck, 1991). According to Spurgeon, Mazelan and Barwell (2012) a long-term exposure to high levels of job-stress may lead to physical and psychological symptoms of distress. Examples of

psychological problems are difficulties with concentrating on tasks, and anxiety, sleeping

problems and depressions. This is in accordance to Boyd, Lewin and Sager (2009), who state that work-stress can have psychological consequences such as emotional exhaustion and job-induced anxiety. Overall, significant evidence is found for the negative effects of job-stress on the health of employees. Besides these observable or perceivable consequences of job-stress, scientific research has shown a positive relation between the amount of felt stress of individuals and their cortisol levels in the morning. People who experience high levels of stress as a result of high workloads, often show high increases in cortisol levels in the first minutes of awaking compared to others who experience less stress. This relationship can be explained by the fact that people with work overloads already start with thinking about how to cope with all the demands and tasks they have to accomplish when they wake up in the morning (Schulz, Kirschbaum, Prüßner & Hellhammer, 1998). On the long-term, a high level of cortisol in one’s blood can lead to several psychiatric problems, such as feelings of anxiety and depression (‘’Cortisol’’, 2013). In general, stress reactions, which are also often called strains, are the consequences of a certain stressor for the well-being of an individual (de Witte, 1999).

The negative consequences of job-stress are not restricted to the individual employees. Bruggen (2015) stated that there is a negative linear relationship between levels of job-stress and job-performance. Also Spurgeon, Mazelan and Barwell (2012) explained that long-term

(10)

10

experience of job-stress reduces the performance on the job since it reduces the interest and efforts of employees. Jamal (2011) conducted a research in Pakistan and Malaysia and found that workers who reported high levels of job-stress generally had low job-performance and workers with low levels of job-stress overall reported higher levels of job-performance. Job-performance is defined as a situation in which an employee is able to perform his or her job demands

successfully, while this employee deals with the normal constraints of the current available resources (Jamal, 1984). Furthermore, Boyd, Lewin and Sager (2009) stated that psychological consequences of job-stress subsequently influence employees’ job-satisfaction and their intention to withdraw. These negative outcomes of job-stress can in the end influence customer

satisfaction via the employees, since employees can deliver lower quality services due to job-stress and revenues can be lost as a result (Boyd, Lewin & Sager, 2009). This makes job-job-stress a very costly problem for both organizations, in terms of financial losses and organizational problems, and for employees, who can experience physical and mental consequences as a result of job-stress.

When employees experience high levels of stress, it is necessary for their managers to think about solutions to bring these stress levels to a lower level. If managers do not intervene, long-term increased levels of stress can lower the performance of employees on the job, bring risks for the employees and for the reputation of their organization, and skilled employees can get lost (Imtiaz & Ahmad, 2009). These negative consequences of job-stress make it clear how important it is that the management of organizations conduct effective practices to attack this problem.

2.3 Reducing and preventing job-stress

According to Niharika and Kiran (2014), it is the responsibility of the management to prevent their employees from experiencing high levels of stress. However, for managers to be able to attack this problem, it is first of all necessary to find out what the drivers of the experienced job-stress are. Once the most significant sources are identified, one can start thinking about which suitable actions management could undertake to reduce the amount of stress of their employees. Actions which decrease the amount of stress, would in the end lead to healthier workers and would at the same time lead to an increase in overall efficiency of the organization and higher performance (Niharika & Kiran, 2014).

(11)

11 2.4 The variety of factors influencing job-stress

Until today the problem of job-stress is unsolved and many researchers have attempted to investigate which factors lead to job-stress. Various theories, models and frameworks have been established over the past decades. The most significant factors, which seem to be highly present in the banking sector will be discussed in this section.

Factors leading to job-stress are also called job-stressors. They can be conditions and events which lead to tensions (Kahn & Byosiere, 1992). The long list of factors which induce job-stress can be categorized into different groups in several ways. First of all, job-stressors can occur in different time spans. Both events which occur once, and states in which employees reside for a longer time, can result in job-stress (Kahn & Byosiere, 1992). Moreover, Spurgeon, Mazelan and Barwell (2012) stated that job-stress can either come from characteristics of the job itself, such as job demands and high workloads, or from based factors such as

role-ambiguity and role-conflict. Furthermore, Beehr and Newman (1978) gave an overview of the different facets of job-stress. They sub-divided these facets in factors from the environment and personal facets. Environmental facets represent job demands and task characteristics, role demands or expectations, organizational characteristics and conditions, and organization’s external demands and conditions. This study will focus on environmental facets, since these are expected to be the factors which managers can influence most easily.

2.5 Workload increases job-stress

Workload is a so called job-stressor, which has often been investigated in previous studies. Workload can be seen as a form of demand stressor (Cordes & Dougherty, 1993) and can be described as the amount of work which has to be accomplished by an individual (Spector & Jex, 1998). To be more specific, one can make a distinction between a quantitative workload and a qualitative workload. Quantitative workload represents the amount of work which has to be accomplished in a specific amount of time, and the qualitative workload has, among other things, to deal with the difficulty of the tasks (Bruggen, 2015).

In a study conducted on the amount of perceived stress among banking employees in India, it was found that workload for a large part determined the perceived work-related stress of employees. In this study, different factors which have already been found to predict job-stress were measured in the Indian banking sector. It became evident that role overload had the highest

(12)

12

overall score (Dhankar, 2015). The quantitative part of this role overload is perceived when employees have the feeling that they have too much work to do in a certain amount of time, which can be seen as a quantitative work overload (Bruggen, 2015). Another study, conducted in Belgium, by van Ruysseveldt et al. (2003) with data gathered in 2001, found that workload was one of the most significant determinants of job-stress amongst banking employees. Besides that, the study also showed that banking employees had higher levels of perceived workload than employees in other sectors. This high perceived workload was expressed by employees during interviews as having the feeling that they have to work hard and fast. Also many employees announced that they work many hours after their normal working day to cope with the high workload placed upon them. Furthermore, these employees said they had the feeling that there were not enough employees left within the organization to fulfill all the daily tasks. These aspects of workload can be categorized as quantitative workload. It became evident that in these previous studies, there was a misfit between the amount of work to be accomplished, and the available capacity of employees. Cobb et al. (1975) also stated that high responsibility loads placed upon employees can lead to job-stress when these individuals are not able to cope with

the increased responsibilities. This can have negative physical and psychological consequences. The list of studies who emphasized the negative relationship between high workloads and

job-stress stress is long. In general, it appears to be the case that when the workload is high, the amount of perceived job-stress is high as well (Bruggen, 2015). In a study conducted by Bruggen (2015) it became clear that workload directly influences the performance of employees. The implication of this finding is that when firms want to reach the optimal performance levels of their employees, they have to create appropriate levels of workload for their employees as well. Workload and the potential job-stress resulting from it can restrict the performance levels of the employees and subsequently of their organizations. This can be explained by the fact that a high workload functions as a barrier which takes the attention from employees away from their tasks (Bruggen, 2015). In this way the performance of employees can decrease as a result of a high workload. Another explanation of the consequences of high workloads for the performance on the job is found by Long, Kahn & Schutz (1992) who found that when employees experience a high workload, their strategies for coping with their demands are mainly emotional and defensive instead of problem-solving behaviour. This makes it harder to perform well on the job. Friedman and Mann (1993) furthermore came to the conclusion that in periods of stress, employees’

(13)

13

thinking is restricted. Larsen (2001) supported these negative effects of stress with stating that in the state of high stress, employees are less able to analyse difficult situations and tasks. These studies show in general that when employees have a high workload, the resulting stress, makes them even less able to cope with these high workloads. This makes the subsequent performance significantly lower, since the amount of work to accomplish is relatively high, but their ability to deal with this high workload has decreased. This effect could be seen as a never ending circle. Interesting is the fact that it has been proven several times in previous studies that a high

workload increases the amount of job-stress, but that this relationship has, to our knowledge, not been investigated in the Dutch banking sector since the most recent recession. Therefore, this paper tries to answer the following sub-question: ‘’Does an increase in perceived workload lead to an increase in perceived job-stress?’’.

In the light of previous findings, it can be expected that high perceived workloads among banking employees lead to high levels of job-stress as well.

Hypothesis 1: ‘’An increase in perceived workload leads to an increase in perceived job-stress’’

2.6 Reorganizations and workload

The banking sector is one of the sectors in which reorganizations occur very often (Tijdens & van Klaveren, 2002). Increased workloads and responsibilities can be the logical result of reorganizations and downsizing efforts. During reorganizations, often a large amount of jobs disappear. As a result, less people are left to accomplish all the tasks (FNV, 2016). On the other hand, employees in the banking sector experience high workloads since they have to perform well in a difficult economic situation (‘’Zeer hoge werkdruk en imagostress in de financiële sector’’, 2012). The combination of employees having the feeling that they have to perform well, despite the difficult financial situation in which their sector resides, with a lesser amount of people left as a result of lay-offs, can be expected to result in high perceived workloads for these employees. The effect of workload, decreases in staff, changes in the organization, long

workdays and too high responsibilities are mentioned by Materson (1980) as causes of job-stress. A logical relationship between downsizing efforts in the banking sector and the perceived

workload can be expected. This relationship is also found in healthcare, where workload has increased significantly after downsizing and reorganization efforts (Nordang, Hall-Lord & Farup, 2010). According to van Ruysseveldt et al. (2003) employees in the banking sector have to work very hard for a relatively insecure job. Employees can get insecure about whether they are

(14)

14

capable of fulfilling all these tasks and accompanying responsibilities which are placed upon them (Cobb et al., 1975). Workload resulting from reorganizations can thus lead to feelings of insecurity for employees, regarding whether they will be able to successfully complete their assigned tasks on time, which make employees feel worried and nervous (Beehr & Bhagat, 1985). These earlier findings in previous studies and the fact that the workload is relatively high in the banking sector, make this factor very interesting to investigate within this sector.

2.7 Job-insecurity increases job-stress

Another factor which is often cited in the literature as inducing stress, is perceived job-insecurity. An interesting fact is that studies have shown that banking employees perceive higher levels of job-insecurity than employees in other sectors and that the experienced levels of stress also are significantly high in this sector (van Ruysseveldt et al., 2003). When employees believe their job-insecurity is high, this means in other words that they fear losing their current job (de Witte, 1999). A study conducted by Harris (2011) showed that job-insecurity is one of the most significant factors in explaining job-stress. Employees in different sectors were asked to rate several factors on the extent to which they predicted the amount of job-stress they felt. Job-insecurity scored significantly high in this ranking of most significant job-stressors.

The effect of job-insecurity on perceived job-stress can be seen in the light of another related job-stressor, namely the locus of control of an individual employee. Spector (1988) states that employees have a specific locus of control, which means that they expect for example promotions, favourable circumstances and salary increases to be under the control of their own actions, or under the control of other factors. When employees perceive their future as controlled by external sources, this means they have an external locus of control. Employees with an

external locus of control seem to attribute their successes or failures to factors outside of their own control. When employees believe that factors outside of their control are influencing their work performance, this often leads to job-anxiety and emotional exhaustion (Spector, 1988). In some way, locus of control can be seen as related to the perceived amount of job-insecurity. When employees personally want to stay at their current organization, but perceive the chance of losing their job as high, due to the threat of getting fired by the organization, they could

potentially perceive a high job-insecurity. It could be expected that employees perceive the decision made by an organization to fire them, as outside of their own direct control.

(15)

15

Many studies have provided evidence for a variety of negative effects of job-insecurity.

According to de Witte (1999) job-insecurity leads to lower general psychological well-being and reduces job-satisfaction. Also job-insecurity leads to more psychosomatic complaints and

physical strains. Roskies, Louis-Guering, and Fournier (1993) found that job-insecurity has an influence on psychological distress and feelings of anxiety and depression. Several other studies furthermore found that job-insecurity leads to higher levels of mental, emotional and physical exhaustion, which is also sometimes called burnout (de Witte, 1999). According to de Witte, this would potentially indicate that when employees are experiencing job-insecurity on the long term, this can use up all of an employee’s resources as for example their energy (Dekker & Schaufeli, 1995).

The serious negative effects of job-insecurity can be explained by the fact that an individual’s job is very important in satisfying certain human needs. When employees feel insecure about their job, this means that they have the feeling that they might lose their current job. First of all, work can include social relationships and recognition by others for one’s work (Jahoda, 1982). Also work involves earning a salary which enables employees to make sure their own and potentially their families’ basic needs are met. Work can also allow individuals to develop themselves. Thus, when employees experience job-insecurity, this might lead to fear of not being able to meet the needs which are satisfied by the job and one can be afraid to lose his or her income and their social contact associated with their job (Jahoda, 1982).

According to de Witte (1999) job-insecurity consists of two aspects. The first aspect is the unpredictability part. This means that it is unknown what will happen in the future. As a result, it is hard to react appropriately and individuals do not know what they should do. A lack of clarity about the future and a lack of clarity about what is expected and which behaviors an employee should adopt, can be seen as unpredictability. According to Warr (1987), this leads to a decrease in the psychological well-being of employees. Interesting is a finding in a study done by Dekker and Schaufeli (1995). In their study it was found that when employees are at a certain point where they are sure that they will lose their job, they perceived less psychological problems than when employees were still insecure about whether they could retain their job or not. The second aspect of job-insecurity is uncontrollability (de Witte, 1999; Vuori, Blonk & Price, 2015). This means that employees perceive to lack the control or power over the threat, which seems to

(16)

16

Even though it is very clear that insecurity can bring serious problems with regard to job-stress, according to de Witte (1999) job-insecurity has received limited attention compared to other variables explaining job-stress. It was often found that job-insecurity has an influence on the amount of stress, but only to a limited extent it has been clear to what extent

insecurity in reality predicts stress in relation to other factors. Besides that, the effects of job-insecurity in the banking sector have only been investigated to a limited extent, mostly in other countries than the Netherlands, as for example in India and before the financial crisis took place (van Ruysseveldt et al., 2003; Niharika and Kiran, 2014). However, it could be expected that job-insecurity has increased since then and relationships between job-job-insecurity and the subsequent perceived job-stress could have changed. One of the questions which this paper tries to answer is: ‘’Does an increase in perceived job-insecurity lead to an increase in perceived job-stress?’’. With the previous literature in mind, it is expected that also in the Dutch banking sector, job-insecurity positively influences the amount of perceived job-stress.

Hypothesis 2: ‘’An increase in perceived insecurity leads to an increase in perceived job-stress’’

2.8 Reorganizations and job-insecurity

Reorganizations are changes in the organization, which can have layoffs or re-positionings as a consequence (FNV, 2016). Tijdens and van Klaveren (2002) conducted a study in the

Netherlands which provided support for a relationship between reorganizations and

job-insecurity. Job-insecurity was perceived as almost three times as high among employees who had experienced a reorganization in the past year as it was among employees who did not experience a reorganization in the previous year. According to Pedraza, de Bustillo and Tijdens (2005), the emergence of new technology and the economic recession lead to structural adjustments within organizations when they adopt downsizing and reorganizations. In the banking sector, both new technologies and economic recession, have had significant impact on the number of employees which are employed (‘’ING verdere digitalisering leidt tot het verlies van 1700 banen’’, 2016; ‘’Banken schrappen ruim half miljoen banen sinds crisis’’, 2016). The way of doing business and daily tasks has changed significantly due to automatization of processes and due to empowering customers by turning tasks which were previously performed by employees, into digitalized self-service tasks for customers. This digitalization has as a consequence that a significant amount of

(17)

17

jobs within the banking sector has disappeared, often during reorganizations (‘’ING verdere digitalisering leidt tot het verlies van 1700 banen’’, 2016). In a study conducted by Tijdens and van Klaveren (2002) it was found that among employees who experienced a reorganization the year before, eleven percent experienced the fear of losing their job, while in comparison only four percent of the employees who did not experience a reorganization was afraid to lose their job. The authors stated that the relationship between reorganizations and dismissals is high. Reorganizations and job-insecurity often go hand in hand. Another significant finding of this study is the fact that in organizations which are highly hierarchical, as well in organizations with highly educated employees, the chance of experiencing a reorganization is higher than in flatter organizations or in organizations with employees with lower levels of education. This makes the banking sector very vulnerable for reorganizations, given the fact that banks are often very hierarchical, while in efforts to react to demands of today’s environment, it is often wishful to make the hierarchies of these organizations flatter (Tijdens & van Klaveren, 2002). These efforts of flattening an organization make that many jobs disappear in this sector. Also fusions of banks lead to higher job-insecurity in this sector (van Ruysseveldt et al., (2004). All these findings make it very interesting to look at the effect of the high levels of job-insecurity in the banking sector on the amount of perceived job-stress.

2.9 Moderating effect of job-insecurity on relationship between workload and job-stress The effect of a high workload on the subsequent amount of perceived job-stress is influenced by the amount of perceived control an employee has over one’s own job (van Ruysseveldt et al., (2004). When employees lack the control over deciding about important aspects of their job, they cannot prevent the negative consequences of high workloads. Employees who experience both a high workload and low control opportunities, face a high risk of experiencing job-stress

according to van Ruysseveldt et al. (2004) This is in accordance with the job demand-control model of Karasek (1979). The control in the demand-control model captures for example

autonomy of individuals to decide upon how they perform the tasks on the job (van Ruysseveldt et al., 2004). This moderating role of the control which an employee perceives to have over his own job in the relationship between workload and the amount of job-stress, makes it interesting to also look at a potential moderating effect of job-insecurity on the relationship between

(18)

18

Given the fact that there is an interaction effect of perceived control and workload together on the experienced job-stress, combined with the idea that job-insecurity could be seen as a form of control over retaining one’s own job, there is a chance of job-insecurity interacting with

workload in determining the perceived stress as well. To illustrate this, one could imagine that an employee who feels insecure about retaining his job, sees the potential consequences of not being able to cope with a high workload as more threatening. They could fear getting fired as a result of not being able to accomplish all the tasks. This worrying could be seen as a form of job-stress. So when people perceive a high level of job-insecurity, the effect of workload on the amount of job-stress might get stronger than it already would be when job-insecurity is lower.

This potential moderating role of job-insecurity on the relationship between workload and job-stress has to our knowledge not been investigated before. However, the general idea of testing a potential moderating effect of insecurity on the relation between a certain job-stressor and the amount of job-stress is not completely new (Kausto, Elo, Lipponen & Elovainio, 2005). In a study conducted by Kausto et al. (2005) it was found that job-insecurity positively moderates the relationship between the job-stressor procedural justice and job-stress. This makes it even more interesting to investigate whether job-insecurity also plays a moderating role in the relationship between workload and job-stress. A positive moderating role of job-insecurity on the relationship between workload and job-stress would mean that the strength of the relationship between workload and job-stress would be positively influenced by the level of job-insecurity. The higher the level of perceived job-insecurity, the higher the increase in job-stress as a result of an increase in workload. This potential moderating role will be investigated by answering the question: ‘’Does insecurity positively moderate the relationship between workload and job-stress?’’.

After reviewing the previous studies mentioned above, hypothesis 3 could be stated.

Hypothesis 3: ‘’Job-insecurity positively moderates the positive relationship between workload and job-stress’’

(19)

19

3. Method

3.1 Research setting

Data was collected with the use of surveys which were sent out to employees at several well-known Dutch banks. The names of these banks were not asked, to ensure that the privacy of the organizations and their reputation would not be threatened. Since the research sample included participants from several banks, these banks were treated as the general Dutch banking sector. No specific characteristics were in place for employees to be selected to participate in this study. 3.2 Data collection

Surveys were used to gather data. These surveys were questionnaires with several statements which were rated on Likert-scales. These statements were indicators of the independent and dependent variables. Control variables were measured with open and closed questions. Surveys were distributed both online and as hard-copy. In the introduction, participants were promised that their information and answers would be treated confidentially and would only be used for the objective of this study. First of all, at one well-known Dutch bank, an agreement on

anonymous treatment of all data was signed during an appointment with a department manager. The manager distributed an e-mail to a group of employees within two different departments. A second way of reaching online participants was via social media, by placing a motivational text together with the survey link on LinkedIn and Facebook. A third attempt to reach a large amount of participants at once, was by sending out e-mails and contacting several bank-offices by phone, to ask them for permission to let their employees participate in this study. Another way of finding potential participants, was by approaching banking employees in areas close to the offices of several well-known banks in Amsterdam South-East. E-mail addresses of employees who agreed with participating in this study were collected and the surveys were distributed to them by e-mail. Lastly, offline data collection was conducted by distributing hard-copy surveys to several local offices of banks in and around Amsterdam.

3.3 Research sample

Data was collected among 98 banking employees at several Dutch banks. Unfortunately, six surveys were not completed, several answers on questions were missing. These cases were excluded from the sample, with a remaining sample size of 92 employees (N=92). This sample consisted of 30 females and 62 males. Their ages varied between 20 years and 61 years old, with

(20)

20

an average of 38 years. The average tenure of employees in the sample was 10 years. Within this sample, 76% of the employees reported that their job was important in securing their family of their needs. Among these employees, 72% has experienced a reorganization in their current organization. The mean job-insecurity score of these employees was 2.4 on a scale of 1-5 which ranges from low to high job-insecurity, which is slightly below the middle point of the scale. Besides that, employees reported an average workload of 3.3 on a scale of 1-5 which ranges from low to high workload. This workload is above the middle level of the scale. Furthermore, the average perceived level of job-stress among employees was 3.1 on a 7-point scale ranging from low to high perceived job-stress. This is slightly below the scale mean, which would indicate that the employees in this sample did not have very high job-stress levels.

3.4 Variables

The dependent and independent variables were measured in the survey with the use of self-reported measures. Employees rated themselves on experiencing the phenomena investigated in this study. The measures of these variables were used in previous studies which have proven their validity and reliability several times before. However, it was still essential to test the reliabilities of these scales in this study as well. The specific reliabilities of the scales used in this research are reported in table 1 between the brackets. These reliabilities were calculated within SPSS with the option ‘’Scale reliabilities’’. A complete overview of the questions asked in the survey can be found in the appendix.

Control variables

Several variables were included in this study as control variables to control for the fact that they would potentially explain a part of the amount of perceived job-stress.

Experienced a reorganization: In this study, it was controlled for whether employees have experienced a reorganization in their current organization or not. This variable is measured by asking participants the question: ‘’Have you ever experienced a reorganization in this

organization?‘’. In previous studies, a positive relationship was found between reorganizations and the perceived job-insecurity of employees (Tijdens and van Klaveren, 2002). Thus,

reorganizations could lead to higher job-insecurity and according to hypothesis 2, job-insecurity would lead to higher levels of experienced job-stress. It is of significant importance to control for

(21)

21

the potential direct influence of reorganizations on the amount of job-stress and to focus on the effect of job-insecurity on job-stress.

Tenure: Tenure was measured in the survey by asking employees how many years they have been working at the current organization. One could imagine that it could be the case that people who have been working at an organization for a longer time, have developed better strategies for coping with stressful situations such as high workloads, than colleagues who have been working in the organization for a shorter time. In a study conducted by Goldenberg and Waddell (1990) among nurses, it was found that the shorter tenure participants had, the higher their perceptions of certain job-stressors were. Nurses with longer tenure were more inclined to apply active coping strategies to job-stressors. On the other hand, in a study by Cheng and Chan (2008) it was found that the relationship between job-insecurity and ill-health was positively moderated by tenure. The explanation of this phenomenon is that individuals with a longer tenure in general are more committed and involved with their job, thus as a result, the fear of losing this job has more negative effects (Cheng & Chan, 2008). Thus, stress-related consequences of certain

job-stressors could differ between employees with different lengths of tenure.

Age: According to Cheng and Chan (2008) age could have an influence on the relationship between job-insecurity and job-stress in the same way as tenure. An important point here is that these researchers stated that age and tenure should be examined separately however. They seem to correlate closely with each other but they are not the same. For this reason, these two control variables were both included in the regression. Tenure and age represent different facets and can both potentially influence the relationship between job-insecurity and ill-health in their own unique way (Cohen, 1991). Tenure captures involvement with a job and organizational commitment (Brown, 1996; Mathieu, Zajac & Appelbaum, 1990). Age on the other hand,

indicates the alternative opportunities for employees on future jobs and their economic insecurity (Kuhnert & Vance, 1992).

Gender: Participants were asked whether they are a male or female. In a study conducted by Jamal (2005), it was found that women in general experienced more negative health

consequences from high workloads, than men. Furthermore, results in a study on cortisol levels showed that individuals who are experiencing stress on the long-term, have larger increases in cortisol levels in the morning than individuals who are not experiencing high levels of stress. A

(22)

22

significant finding was that women in general experience higher increases in cortisol in the morning as a result of long-term exposure to stress than men (Schulz et al., 1998). This makes it interesting to look at potential differences between women and men in the effects of workload and job-insecurity on the amount of felt stress. It could be the case that for women, the

influences of certain job-stressors are stronger than for men.

Importance of job in securing family: Participants were asked whether their job is important in securing their family. The anxiety and worry resulting from a high perceived job-insecurity could be especially high for employees whose income is the basis for securing their family in their needs. In a study conducted by Kinnunen, Mauno, Natti, & Happonen (1999), the extent to which employees were dependent on their job, partly determined the consequences of job-insecurity.

Dependent variable

Perceived job-stress: The amount of job-stress perceived by employees was measured with four questions which were also used in a study by Netemeyer, Maxham and Pullig (2005). The measure consisted of four items. Three of these items were introduced by House and Rizzo (1972) and one item was developed and added by Netemeyer, Maxham and Pullig (2005). This measure indicates the amount of perceived stress with four statements which are rated on a 7-point Likert-scale ranging from 1 (disagree strongly) to 7 (agree strongly). An example of these statements is ‘’At the end of the day my job leaves me stressed-out’’. The ratings of individual participants on these four statements were summed up and divided by four to represent the overall job-stress score of each individual. A high score on these statements indicates a high level of job-stress perceived by an employee. Cronbach’s alpha of this measure has proven to be high in earlier research, which means the four items have satisfying internal consistency and that this scale is a reliable measure of job-stress. In the current study, Cronbach’s alpha was

measured once again to make sure a sufficient reliability was also found among the employees in this sample. The reliability of the scale was found to be highly reliable (a = 0.87).

Independent variables

Perceived job-insecurity: Job-insecurity was measured with a list of questions which were also applied in a study by Pienaar, de Witte, Hellgren and Sverke (2013). This list of questions is a shortened version of the original measure developed by de Witte (2000) and contains questions

(23)

23

to measure employees’ job-insecurity level. The currently used list contains four cognitive items as well as four affective items. The complete overview of these eight questions can be found in the appendix. These questions were rated on a 5-point Likert-scale which ranges from 1

(disagree strongly) to 5 (agree strongly). An example of a question is ‘’I am very secure that I will be able to keep my job’’. In the survey, the first four of these items were stated so that a high rate would indicate a low job-insecurity and the four remaining items were stated the other way around. Thus, the first four items were reverse recoded after data collection, so that in the end a high rate on each of the eight items indicated a high job-insecurity. Rates of each participant were averaged to create an overall job-insecurity score for each participant. A high score on this measure indicates a high level of perceived job-insecurity for this participant. The reliability for both the cognitive and the affective items were measured in earlier studies and Cronbach’s alpha’s were highly sufficient for both kind of items. In the current study, the reliability of this scale was confirmed to be highly reliable as well (a =0.90). Thus, one could conclude that these eight questions could be seen as reliable measure and could be applied in this research to

accurately measure the level of perceived job-insecurity of an employee.

Perceived workload: The perceived workload of employees was measured with questions which were used and validated in a research conducted by Spector and Jex (1998). This measure

contains five items which ask employees how often certain situations occur on their job. Overall, these questions ask employees how often they are forced to work very hard. The measure gives an indication of the amount of work and the rate with which it has to be accomplished (Spector & Jex, 1998). The statements were rated on a 5-point Likert-scale which goes from 1 (less than once per month or never) to 5 (several times per day). One of the statements was ‘’How often does your job require you to work very fast?’’ Answers of employees on these items were averaged. A high score on these items indicates a high perceived workload. In the original study of Spector and Jex (1998) Cronbach’s alpha was found to be high. In this current study the reliability was also found to be more than sufficient (a = 0.86). The measure is a shortened version of the original scale, as it was developed by Spector (1987). The original scale contained eight items, which were both quantitative items and qualitative items of workload. However, the authors decided to shorten the list of statements to generate this version of five items since this would lead to an increase in internal consistency. This measure is an indicator of the amount of

(24)

24

perceived quantitative workload. As a result of choosing to focus on the quantitative workload in this study and of subsequently using this quantitative measure, the qualitative workload is out of

the reach of this study.

In general, one could conclude that the measures of workload, insecurity and job-stress used in this study all have a satisfying scale reliability of 0.8 or higher, which means they are highly reliable. In other words, the combinations of the statements in the survey accurately and consistently measured what they were intended to measure.

4. Results

4.1 Correlations

In this study, both the main effects of perceived workload and job-insecurity on the amount of perceived job-stress were measured, and the potential moderating effect of job-insecurity on the relationship between workload and job-stress. Before these relationships were investigated into detail by conducting a regression analysis, correlations were calculated between all variables to get a first impression of which relations were present in this study. The resulting Pearson correlation coefficients can be found in the table on the next page. In addition to correlations, also the mean scores of individuals in this study on the variables and the standard deviations are reported. Furthermore, the table shows the reliability of the used measures for the variables workload, job-insecurity and job-stress on the diagonal between brackets.

(25)

25

Table 1. Descriptives and correlations between the variables (Cronbach’s alphas for the dependent and independent variables between brackets)

Note. N= 92. *p<0.05. **p<0.01.

Pearson’s correlation coefficients between all the variables could be interpreted as a first indication to see which relations exist between variables. An interesting correlation was the significant positive correlation between workload and job-stress, r (92) = 0.37, p =0.00. This supports the first hypothesis which stated that when workload increases, job-stress also increases as a result. No significant correlation was found between job-insecurity and job-stress,

r (92) = 0.09, p =0.38. This finding is not in line with the expectation regarding hypothesis 2. Further analysis was conducted with the goal of exploring these findings into more detail. Another interesting, strong positive correlation was found between age and tenure, r (92) = 0.73, p=0.00. Furthermore, a significant positive correlation was found between the independent variables job-insecurity and workload, r (92) = 0.26, p = 0.01. Moreover, the control variables age, tenure, whether an employee’s job is important in securing one’s family and whether an employee has experienced a reorganization in the current organization, do all significantly positively correlate with the independent variables workload and job-insecurity. Within the 1 Job-stress 3.1141 1.36348 (0.866) 2 Workload 3.2891 .88423 .368** (0.858) 3 Job-insecurity 2.394022 .8216538 .093 .260* (0,901) 4 Age 37.76 10.826 .041 .312** .242* 5 Tenure 9.8508 9.32008 .037 .317** .233* .725** 6 Female .33 .471 -.016 .088 .104 -.013 .060 7 Secures family .7609 .42889 .174 .271** .251* .454** .332** .064 8 Experienced reorg. .7174 .45273 -.098 .217* .240* .372** .452** -.027 .157

(26)

26

control variables, age and tenure significantly and positively correlate with the fact whether an employee has experienced a reorganization and whether an individual’s job is important in securing his or her family. Overall, one could state that all control variables, except gender, seem to correlate significantly and positively with the independent variables workload and

job-insecurity. However, none of the control variables seemed to correlate significantly with the outcome variable job-stress.

The exact relationships between the control, independent and dependent variables were further investigated in relation to each other in a regression analysis. The high correlations between job-insecurity and workload, could lead to worrying about multi-collinearity. To check whether there were serious problems of collinearity in the planned regressions, multi-collinearity tests were performed. It was found that none of the variables included in the regressions had a VIF value above 2.5, which indicated that there was no need to be concerned about multi-collinearity problems between the variables. Thus, after these initial tests have shown that the scales of these independent and dependent variables are reliable, and there were no problems with regard to multi-collinearity, further analysis regarding the hypotheses could be done with the use of linear regression analysis.

4.2 Regression analyses

The three hypotheses were tested in SPSS with hierarchical linear regressions on the variables to find out how much variance in job-stress would be explained by workload and job-insecurity and by a potential moderating effect of job-insecurity. The regression was build up in three steps, resulting in a final regression which regressed job-stress on the control variables, independent variables and an interaction term. These variables were added to the regression model in this order as well. In the first step only the control variables were included in the regression. In the second step the independent variables workload and job-insecurity were added to investigate to what extent they explained the amount of job-stress. In other words, the second step tested the main effects of workload and job-insecurity on job-stress, controlled for the control variables age, gender, tenure, experienced a reorganization and job secures family. In the third step, a potential moderating effect was tested by adding an interaction term of workload and job-insecurity. In other words, the third step checked whether workload*job-insecurity predicts the

(27)

27

amount of stress to find out whether a moderating effect exists. To investigate whether job-insecurity moderates the relationship between workload and job-stress, the values of R-squared resulting from the regression models within Step 2 and 3 were compared on the proportion of variance explained.

To make sure that no problems would arise with the fact that job-insecurity and workload are significantly correlated, these variables were mean centered before conducting regression analyses. Mean centering the variables which are included in an interaction term is strongly recommended by Cohen, Cohen, West and Aiken (2003). This was done by subtracting the sample mean score on job-insecurity and workload from the individual score of each participant. To create an interaction term of workload and job-insecurity, these two mean centered values of job-insecurity and workload were multiplied with each other. For the resulting regression analyses to give accurate values of the job-stress score which were easy to interpret, also the continuous control variables ‘’age’’ and ‘’tenure’’ were mean centered, by subtracting the overall sample mean age and tenure from each individual employees’ specific tenure and age. Mean centering all continuous variables instead of mean centering only the variables which are included in the interaction term was chosen for due to strong recommendations of Cohen et al. (2003) to center all predictors in a regression before conducting further regression analyses. The variables which are represented in table 2 with results, are thus mean centered except for the dichotomous control variables (whether employees experienced a reorganization, whether their job is important in securing their family and whether they are a female or not). These

dichotomous control variables were recoded to the values 0 and 1. For gender, a 1 would mean female and 0 would mean male. Employees who experienced a reorganization at their current organization received a value of 1 on this item and a 0 if they had not experienced a

reorganization at their current employer. Lastly, employees whose jobs are important in securing their family, have a score of 1 on this item, while employees whose job does not play a role in securing a family scored a 0 on this item. Furthermore, the dependent variable job-stress was also not centralized.

The results of the linear regression which was conducted by taking the three steps can be found in the table below. In this table, one can find the standardized Beta coefficients, values of R-squared and F-change.

(28)

28

The interpretations of these values will be explained in specific sections assigned to the tested hypotheses.

Table 2. Regression of perceived job-stress on age, tenure, female, experienced a reorganization, job secures family, perceived workload and perceived job-insecurity

Perceived job-stress

Step 1 Step 2 Step 3

Age -.050 -.092 -.065

Tenure .077 .018 -.005

Female -.037 -.069 -.041

Exp. Reorg. -.147 -.187 -.201

Job sec. fam. .197 .129 .112 R² .050 Workload .395*** .431*** Job-insecurity .029 .052 R² (F-change) .187 (7.074) Workload*Job-insecurity -.186 R² (F-change) .218 (3.318) Note. N= 92. *p≤0.05. **p≤0.01. ***p≤0.001. 4.3 Regression results regarding hypotheses 1 & 2

Hypothesis 1 stated that an increase in perceived workload would lead to a subsequent increase in perceived job-stress while hypothesis 2 stated that an increase in perceived job-insecurity would lead to an increase in perceived job-stress. As explained above, in Step 1 of the regression analysis, job-stress was regressed on the control variables age, tenure, gender, experienced a reorganization or not and job secures family. In Step 2, the independent variables workload and job-insecurity were added to the regression to test their independent main effects on job-stress.

(29)

29

As a result, in this step the regression included the control variables and the independent variables workload and job-insecurity, to test the two hypotheses regarding the main effects of workload and job-insecurity.

In the regression results it became evident that none of the control variables was a significant predictor of job-stress on itself, even though one variable was close to significant. In this study a significance level of P<0.05 is applied. However, in several other studies,

significance levels of P< 0.1 are applied. The control variable ‘’experienced a reorganization’’, β= -0.19, p = 0.1 was close to this significance level of p<0.1. Furthermore, the values of R-squared show that the control variables and the independent variables workload and

job-insecurity together explained 18.7 percent of the total variance in job-stress. The only variable in Step 2 which significantly predicted job-stress was workload, β =0.4, p =0.00. A remarkable finding was that the variable insecurity was not significant in predicting the amount of job-stress, β =0.03 and p =0.79, which was against the expectations. The high p-value of job-insecurity in the regression shows that job-job-insecurity does not have a significant main effect on job-stress. The p-value of F-change as a result of including the control variables in the first step, which was p =0.48, shows that the control variables did not lead to significant explanation of variance in the dependent variable job-stress. However, together these control variables

explained 5 percent of the variance in job-stress. The addition of the two independent variables workload and job-insecurity in the regression in Step 2 has a p-value of F-change of p =0.00. The increase in the percentage of explained variance in job-stress as a result of adding the variables workload and job-insecurity was 13.7 percent. From the results above, one can

conclude that the predicted positive main effect of workload on job-stress is found, in contrast to the fact that no significant main effect of job-insecurity on job-stress was found. In other words, hypothesis 1 which stated that an increase in workload leads to an increase in job-stress was confirmed. Hypothesis 2, which stated that an increase in job-insecurity leads to an increase in job-stress, was not supported in this study.

4.4 Regression results regarding hypothesis 3

Hypothesis 3 predicted that job-insecurity would positively moderate the relationship between workload and job-stress, such that for employees with a high perceived job-insecurity, the

(30)

30

feel secure about retaining their current job. This hypothesis was tested within Step 3. In this step, it was expected to find a significant positive interaction effect of workload and job-insecurity together on job-stress. Interaction effects mean that the effect of an independent variable on the outcome variable depends on the level of a third variable, which is called a moderator variable. In line with hypothesis 3, it was expected that the regression in Step 3, with the interaction term included, would significantly predict the amount of job-stress more

accurately than the regression in the previous Step 2, without this interaction term. In specific terms, it was expected that the value of R-squared resulting from Step 3, with interaction term included would be significantly higher than the value of R-squared of Step 2 with only the independent variables and control variables included. The results of the regression analysis can be found in table 2 on page 28.

The regression led to interesting results. First of all, the interaction effect has a negative standardized Beta, β = -0.19 which was against the expectations. However, this negative interaction effect was not significant, p = 0.07. The proportion of variance explained in the regression with interaction term included is 21.8 percent in comparison with an explained variance of 18.7 percent in the regression without interaction term included (see table 2). The increase in explained variance of job-stress as a result of adding the interaction term was thus 3.1 percent. The interaction does explain some variance in the amount of job-stress, but this

improvement in predictive power as a result of adding the interaction term is not significant. However, this p-value of F-change is close to the applied significance level of p<0.05, since p=0.07. This makes it interesting to investigate into more detail what really happens in the relationship between the variables.

To investigate how this almost marginally significant, negative moderating effect could be interpreted, a simple slope analysis was performed in the program ‘’Process’’ of SPSS. Cohen et al. (2003) recommended to investigate interactions with the use of this tool. With this

program, the relationship of workload and the subsequent amount of perceived job-stress can be compared between cases with low and high levels of the moderator job-insecurity. Within Process, the high and low values of the moderator variable are selected as one standard deviation above and below the average value of the moderator within the sample. From the output, it became clear that the effect of workload on stress for people with a relatively low

Referenties

GERELATEERDE DOCUMENTEN

Moving on to the results of the moderated mediations it was expected that a rater who has a high self-perceived attractiveness would find him or her self similar to an attractive

Although the comprehensive approach lays the guidelines for interaction between military and civilian actors, the amount of civilian actors currently participating in

Because of these motivations, China does not take into consideration issues of governance or political stability of the recipient nation, and thus the results of Chinese

Sure, you could go to the terms of conquer, invade, colonize, but where is the image that persists? The speech is that this territory is concerned because Bolivia does not meet the

Appendix II: Articles selected for discourse analysis This appendix presents an overview of the qualitative sample that is used for the discourse analysis that looks into the

India’s foreign policy tradeoffs: internal politics, borders, regions &amp; human rights The federal design in India and the (vested) interests in Tamil Nadu’s fishing industry, have

How does access to different social capital (informal groups) and human capital (skills) contribute to the self-perceived employability of millennials in

The estimate for the coefficient with a lag of three year can be explained on the following way: In a country with one million inhabitants, where in a certain year one